The negative consequences of substance use among college students are a national public health concern. Prior research has identified risk factors that contribute to substance use, but a gap exists in translating this knowledge into interventions to prevent substance-use related harm. The pervasive use of smartphones with social networking, location-based, and other applications (apps) offers new possibilities for prevention efforts that can be delivered at the precise time when students are making decisions about substance use and related risks. Our long-term goal is to develop, implement, and evaluate mobile approaches that harness these technologies to support college students in avoiding substance use, reducing substance use, and avoiding substance-related harm. A critical first step is to understand how mobile technologies are used in real-time during episodes of alcohol, marijuana, and other drug use, so interventions that fit with existing patterns of technology use can be developed. The two objectives of this application are (1) to apply the Haddon Matrix, commonly used by engineers in the analysis and prevention of injuries, to assess individual, social, environmental, and technological factors during three phases (planning, actively using, and leaving) in substance use episodes, and (2) to use this detailed understanding of substance use episodes to identify practical, acceptable, and feasible mobile interventions targeted to the intersections of technology use with modifiable risk factors. The proposed study will recruit a representative sample of 300 college students at the University of Massachusetts-Amherst and the University of Washington who report using alcohol, marijuana, or other drugs in the past 7 days. In-person interviews will involve measurement of the quantity and frequency of substance use in the prior 7 days and a comprehensive assessment of the timing, location, social setting, and environmental context for the highest-quantity episode. For this episode, technology use data will be obtained via an innovative, participant-led review of his/her own mobile communication, including time-stamped messages. This data will permit identification of mobile intervention ideas targeted to particular substances and settings. These ideas will be presented to a subsample of students in focus groups designed to solicit feedback on feasibility and acceptability. Through this process, we will refine ideas for targeted mobile interventions to prevent harm from substance use that integrate easily into students' existing life routines. The research team of experts in new media technology, college student substance use, and injury prevention will use the results from this study to inform the development of mobile interventions that will be tested in future grant applications. Given the popularity and ubiquity of mobile devices among the college population, and the high incidence of substance-related harm, the interventions that stem from this proposal have potential to lead to widespread improvements in college student health.